LeoZ wrote:whats wrong with a single blind? or, less complicated, just tasting shots from the 2 machine types!

Part of the problem is the unavailability of two espresso machines that are representative of the HX/double boiler genre and are otherwise identical. The good news is that the Vibiemme Domobar Super and its new double boiler sibling could be used for such a side-by-side comparison.

LeoZ wrote:all a double blind in espresso taste tests would do is slow the barista down. without knowing the beans, the barista would have to take that many more shots to adjust for the unknown.
whats wrong with a single blind? or, less complicated, just tasting shots from the 2 machine types! sounds like that hasnt even been done based on your comment above.

Double blinding removes unconscious bias on the part of the person conducting the experiment. Yes, I think it's critical, and no, it shouldn't slow things down. In a taste test like this, double blinding adds a minimal level of extra effort. Instead of the barista serving drinks directly, the cups would be labelled (typically with a random number), and then a different person (who didn't know which cup was which) would actually serve them to the tasters.

HB wrote:Part of the problem is the unavailability of two espresso machines that are representative of the HX/double boiler genre and are otherwise identical. The good news is that the Vibiemme Domobar Super and its new double boiler sibling could be used for such a side-by-side comparison.

That's right. You'd want to remove as many complicating factors as possible. And it would be nice to try a variety of beans, brew ratios, temperatures, etc. It's not easy to do good science...

RapidCoffee wrote:Double blinding removes unconscious bias on the part of the person conducting the experiment. Yes, I think it's critical, and no, it shouldn't slow things down. In a taste test like this, double blinding adds a minimal level of extra effort. Instead of the barista serving drinks directly, the cups would be labelled (typically with a random number), and then a different person (who didn't know which cup was which) would actually serve them to the tasters.

i am assuming you are including the barista as part of the blinding process. difficult to do, since it would take all of 10 seconds for the barista to figure out which machine is which. if you are, thats more of a triple blind test.

if you are not, then you are in fact giving the barista the chance to learn each machine and make it work best, which was my original point.

if you dont want the barista to hand the drinks over b/c a timed wink would influence the taster, then i understand your point there.

HB wrote:Part of the problem is the unavailability of two espresso machines that are representative of the HX/double boiler genre and are otherwise identical. The good news is that the Vibiemme Domobar Super and its new double boiler sibling could be used for such a side-by-side comparison.

but that didnt prevent a taste test of the full manual lever machine vs the spring assisted one....

LeoZ wrote:i am assuming you are including the barista as part of the blinding process. difficult to do, since it would take all of 10 seconds for the barista to figure out which machine is which. if you are, thats more of a triple blind test.

if you are not, then you are in fact giving the barista the chance to learn each machine and make it work best, which was my original point.

if you dont want the barista to hand the drinks over b/c a timed wink would influence the taster, then i understand your point there.

The barista needs to pull the shot properly, and that implies a good working knowledge of the equipment at hand. We have to assume that participants act in good faith, and do not compromise the experiment by winking or giving other signals. So why double blind? Even subtle subconscious clues can influence the outcome of an experiment, especially when testing something as subjective as taste.

Triple blinding is less common, and extends double blind testing to the data analysis phase, not the data gathering procedure.

RapidCoffee wrote:The barista needs to pull the shot properly, and that implies a good working knowledge of the equipment at hand. We have to assume that participants act in good faith, and do not compromise the experiment by winking or giving other signals. So why double blind? Even subtle subconscious clues can influence the outcome of an experiment, especially when testing something as subjective as taste.

Triple blinding is less common, and extends double blind testing to the data analysis phase, not the data gathering procedure.

we are on the same page. i assumed by double blind you meant barista and taster..
now, i just have to buy yet another machine

LeoZ wrote:but that didnt prevent a taste test of the full manual lever machine vs the spring assisted one....

In that case, the difference was significant enough that we could assert the brew pressure profile was the major contributor, even though the two machines were not otherwise identical. In the case of flat versus humped brew temperature profile, the differences have not been shown to be significant by clear consensus. For example, Abe's Buyer's Guide to the Expobar Brewtus conclusion did not assert the Brewtus taste profile was superior to the Giotto, only that it was more consistent ("The key words in the 'Brewtus advantage' are: ease, consistency, and repeatability.").

Since the difference is arguably small, the only way to convincingly argue one way or another is isolation of the variable in question and lots of double blind taste tastes. Ken Fox's Coffee: To Freeze or Not to Freeze demonstrates just how much time and planning such an effort requires.

HB wrote:In that case, the difference was significant enough that we could assert the brew pressure profile was the major contributor, even though the two machines were not otherwise identical. In the case of flat versus humped brew temperature profile, the differences have not been shown to be significant by clear consensus. For example, Abe's Buyer's Guide to the Expobar Brewtus conclusion did not assert the Brewtus taste profile was superior to the Giotto, only that it was more consistent ("The key words in the 'Brewtus advantage' are: ease, consistency, and repeatability.").

Since the difference is arguably small, the only way to convincingly argue one way or another is isolation of the variable in question and lots of double blind taste tastes. Ken Fox's Coffee: To Freeze or Not to Freeze demonstrates just how much time and planning such an effort requires.

again, it seems to come down to the expense of time. hats off to ya guys.

anyway, you are assuming that the 2 VB machines, outside of pump or boiler setup, are otherwise similar. safe assumption?

if not, i think a mystery remains. it will come down to results from different manufacturers and their taste profile design from the factory, with our mouths as the judge to what machine each person likes best..